Gendered Challenges in Academia: Exploring the Impact of Working Hours, Stress, and Job Satisfaction Among Mid-Level University Staff in Germany

Abstract

This study examines the relationships between job satisfaction, overtime hours, perceived stressors, and burnout symptoms among academic mid-level staff at German universities, with a particular focus on gender differences. Drawing on survey data from 1442 academics collected in April/May 2023, this study applies t-tests and regression analyses to examine the effects of structural and personal factors on job satisfaction. The findings suggest that job satisfaction is primarily shaped by psychosocial and institutional conditions. Negative predictors are perceived job insecurity, burnout symptoms, and excessive overtime, whereas a strong dedication to work buffers against these. Variables such as gender, age, parenthood, and participation in structured PhD programs did not show substantial effects. Notably, respondents who postponed having children for professional reasons reported lower job satisfaction, pointing to potential conflicts between career and personal life expectations.

1. Introduction

Job satisfaction, often correlated with job stress and work engagement (Mudrák et al., 2018), is one of the strongest predictors of occupational well-being, job-related productivity, and overall mental health (Cao et al., 2022). These links are particularly consequential in high-pressure environments, such as universities, where job satisfaction directly shapes retention and career advancement (Shen & Slater, 2021).

A selection of studies has so far dealt with work-related stress and satisfaction. Both faculty (Fetherston et al., 2020; Mudrák et al., 2018) and students (Barbayannis et al., 2022) report elevated stress levels, burnout, anxiety, and depression. Stress is often used as a variable that is in fact operationalized using a wide variety of factors (stressors), including high workload, research and teaching pressure, administrative tasks, and work–life conflicts (Lesener & Gusy, 2017; Urbina-Garcia, 2020).

These stressors are felt most acutely by academic staff below the professorial rank (the German “Mittelbau”, Engl. mid-level academics). In this group, unregulated working hours, publication pressure, and a dominance of fixed-term contracts raise concerns about job satisfaction, well-being, and long-term career prospects (Ambrasat, 2019; Briedis et al., 2020; Gleirscher et al., 2022; Lesener & Gusy, 2017). Perceived contract uncertainty has often been mentioned as a heavy burden (Briedis et al., 2020) but has rarely been treated as a distinct stressor.

Stress and satisfaction levels also vary with sociodemographic characteristics. Female academics frequently report higher stress and poorer mental-health outcomes than their male colleagues (Corvino et al., 2022; Shen & Slater, 2021). Key drivers include caregiving responsibilities, gendered role expectations, and unequal advancement opportunities (Ysseldyk et al., 2019). However, looking at the body of research to date that examines gender differences in relation to stress and job satisfaction, the results are ambiguous, which is why further research is needed.

Taken together, these streams in the literature reveal two unresolved questions: (a) which structural and personal stressors most strongly predict job satisfaction under conditions of contractual precarity, and (b) whether these patterns differ by gender.

The present study provides empirically grounded insights into patterns and predictors of job satisfaction among academic mid-level staff at German universities. Using a large-scale survey, we examine structural and personal stressors (e.g., working hours, perceived job insecurity, work dedication) and analyze their association with job satisfaction across disciplines and demographic groups, with a special focus on gender. The aim is to inform institutional policies and support structures in higher education. This empirical approach addresses the need for broader, generalizable data on working conditions in academia, which remains a gap in the current research. The research questions include the following:

• In what ways do work-related variables differ between male and female1 academics occupying mid-level positions within German academia?
• Which organizational and individual factors influence job satisfaction among academic staff?

2. Current State of Research

A growing body of research addresses well-being and job satisfaction in academia across global higher education systems. However, research focusing specifically on gender differences in academic job satisfaction remains limited, particularly within the German context.

Urbina-Garcia (2020) conducted a systematic review of 28 international studies on the mental health of academic staff. The findings revealed consistently high levels of stress and burnout and low levels of well-being. Key stressors include excessive workload, job insecurity, administrative burdens, and insufficient organizational support. Gender disparities were scarcely addressed; only one study (Slišković & Maslać Seršić, 2011) reported that women experienced higher stress levels than men.

Within Germany, Lesener and Gusy (2017) synthesized 43 empirical studies on work conditions and health indicators in academia, using the Job-Demands-Resources (JDR) model as their framework. This model identifies both negative factors (job demands) causing burnout and positive factors (job resources) that reduce burnout and enhance engagement at the same time. Job demands include, among other factors, work overload and future job insecurity. Job resources include support from others, job control, and performance feedback (Schaufeli, 2017) as well as self-efficacy as central predictors for job satisfaction and burnout (Xanthopoulou et al., 2007). Within this framework, Lesener and Gusy (2017) identified three primary stressors in their review: (1) fixed-term contracts and resulting job insecurity, (2) time demands compensated by overtime, and (3) work–life incompatibility. Key resources included creative autonomy, collegial support, and task content. Despite the model’s holistic scope, the review notes that gender differences were largely ignored in the included studies.

While the JDR framework helps to organize prior evidence, our study adopts a data-driven approach rather than testing specific JDR pathways. Taken together, the findings of Urbina-Garcia (2020) and Lesener and Gusy (2017) point to job satisfaction as an important predictor for academic output and six variables that are central to understanding working conditions among mid-level academic staff. We therefore structured our research around these variables and their gender disparities and present the relevant research in the following paragraphs:

• Job satisfaction in academia (as dependent variable);
• Work dedication;
• Self-efficacy;
• Burnout;
• Perceived job insecurity;
• Overtime work;
• Coping strategies.

Job Satisfaction (“the extent to which people like (satisfaction) or dislike (dissatisfaction) their jobs” (Spector, 1997, p. 2))” is often studied as one central dimension of overall faculty well-being because it is associated with better productivity (Cao et al., 2022). Mudrák et al. (2018) operationalized well-being using job satisfaction, stress, and work engagement. Their study among academic staff in Czech universities (n = 1389) supports the JDR model. While average satisfaction scores do not differ between men and women, there are some gender-related differences in predictors for job satisfaction, e.g., perceived job insecurity (see below). Frei and Grund (2020) found that male academics in Germany report significantly higher job satisfaction compared to female academics. Fontinha et al. (2019) observed a similar situation in their UK sample. In contrast, Fetherston et al. (2020) found that male academics in Australia had significantly lower satisfaction scores than their female counterparts. These findings suggest that gender disparities in job satisfaction are context-dependent and need further investigation.

Work dedication—the enthusiasm-and-identification facet of engagement—merits a separate focus because it is the engagement element most strongly linked to job satisfaction and downstream economic gains (Demerouti & Bakker, 2006; Schaufeli et al., 2002). Most studies, however, only report overall engagement. These show that supportive climates raise engagement (Barkhuizen et al., 2014; Hakanen et al., 2006) and heavy demands lower it (Rothmann & Jordaan, 2006); higher engagement, in turn, reliably predicts job satisfaction across sectors, including higher education (Schaufeli et al., 2002; Mudrák et al., 2018). In a Portuguese university, engagement lifted satisfaction for both sexes but more so for women (Mascarenhas et al., 2022), while among German postdocs, over-commitment increased strain mainly for women and fueled exit intentions (Dorenkamp & Weiß, 2018). When dedication itself is reported, academics score high (Barkhuizen et al., 2014; Bezuidenhout & Cilliers, 2010; Mudrák et al., 2018), yet gender is seldom analyzed and findings are mixed. For example, Bezuidenhout and Cilliers (2010) studied a female-only sample that showed low dedication owing to high cynicism, leaving no male comparison. Thus, despite dedication’s central role, gender-segmented evidence, especially from German higher education institutions (HEIs), is scarce. Our study therefore targets dedication rather than the broader engagement construct to fill this gap.

Self-Efficacy (one’s belief in their capability to successfully perform a task that influences their life in some way (Bandura, 2008)) is consistently identified as a key personal resource in the JDR model and was found to be a reliant predictor for job satisfaction in various studies as it buffers against burnout and enhances job engagement (e.g., Chan et al., 2020; Lesener & Gusy, 2017; Schaufeli, 2017). While some studies (Ismayilova & Klassen, 2019; Mudrák et al., 2018) have reported gender-neutral outcomes, others have highlighted disparities, often favoring men, especially when it comes to researching self-efficacy (Epstein & Fischer, 2017; Hemmings & Kay, 2009; Schoen & Wincour, 1988). Hall et al. (2019) conducted a longitudinal cross-national survey (n = 3071) with post-secondary staff in 69 countries investigating on the relationship between self-efficacy and burnout. They found that higher levels of self-efficacy at the baseline correlate with lower levels of burnout during the course of the study but could not find gender disparities. Though they did not investigate correlations with job satisfaction, the results are relevant as burnout has proven to be a strong predictor for job satisfaction in many studies.

Burnout—defined as the chronic depletion of emotional and physical resources in response to prolonged work stress (Maslach, 1993)—remains a pervasive risk for mid-level academics, and the evidence suggests important gendered nuances. In South Africa, Bezuidenhout and Cilliers (2010) reported moderate burnout levels among female academics, marked by emotional exhaustion and growing cynicism even in the presence of relatively high work-engagement scores. This study was conducted among females only. A study in Nigeria could not detect gender-related differences in burnout but measured that women more often feel reduced personal accomplishment, a factor loading on burnout (Adekola, 2010). Extending these insights, Ysseldyk et al. (2019) used a mixed-methods design with postdoctoral women in North America and Europe: survey data confirmed moderate stress, depressive symptoms, and low life satisfaction, while interviews revealed that career uncertainty, limited control, and identity tensions intensified burnout risk. Again, this study was conducted among female researchers only. In a broader faculty sample among female and male scientists, Catano et al. (2010) found that Canadian women faculty reported more physical and psychological health problems than men, and Shen and Slater (2021) showed that pandemic-related occupational stress eroded emotional well-being among academics, although they observed no significant gender differences in that smaller Northern Irish sample. German data are still scarce, but Geier et al. (2025) highlighted that women—especially those working part-time or with caregiving duties—report higher perceived workload pressure and more burnout symptoms than their male colleagues, despite accruing fewer overtime hours. Further studies around the globe support these ambivalent results, e.g., in Brazil (Rocha et al., 2020) and Japan (Taka et al., 2016). Both studies find either small or non-significant gender gaps, reinforcing that gender-related burnout effects vary by context. Gender-focused research and post-pandemic evidence from German HEIs remains notably scarce. Collectively, these studies indicate that burnout among academic staff is not solely a function of workload; it intertwines with perceived job insecurity and gendered role expectations, all of which shape the trajectory from stress to exhaustion and ultimately influence overall job satisfaction.

Job Insecurity is a widespread concern in academia, often linked to fixed-term contracts and uncertain career trajectories that negatively affect job satisfaction (Dorenkamp & Süß, 2017; Frei & Grund, 2020; Geier et al., 2025). While Dorenkamp and Süß (2017) did not observe gender-specific differences in their German sample, Mudrák et al. (2018) observed that male faculty in Czech universities perceived less job insecurity than women, even after controlling for leadership roles. Similar patterns were observed by Corvino et al. (2022) in Italian PhD students, where women reported lower career development prospects and job autonomy. According to the authors, these findings reflect the structural weaknesses in the Italian academic system, particularly around unclear career paths, limited research funding, and implicit gender biases in task allocation. Apart from Dorenkamp and Süß (2017), there are no gender-specific investigations on the influence of job insecurity in German HEIs, although 81% of academic mid-level staff at universities and 63% at universities of applied sciences in Germany are on fixed-term contracts (Gassmann et al., 2025). In Spain, female early career researchers report significantly more cognitive effort to reconciling career and future parenthood due to reportedly higher barriers in establishing a stable career after giving birth, causing them higher levels of concern (Bonache et al., 2020).

Overtime Work and blurred work–life boundaries are key stressors in academic life (e.g., Ambrasat, 2019; Frei & Grund, 2020; Lee et al., 2022). Fetherston et al. (2020) linked excessive working hours and intrusive work-related thoughts to lower well-being. Geier et al. (2025) reported a gendered distribution of overtime: men work more overtime than women, but women experience higher cognitive stress and burnout symptoms, especially those with caregiving duties. Frei and Grund (2020) found similar results, reporting only women with caregiving duties as working less overtime. This suggests that men’s overtime may be more structurally supported, while women’s stress is exacerbated by multitasking and emotional labor.

Together, these findings underscore the need for more gender-sensitive, context-aware research on job satisfaction in academia. While individual studies provide valuable insights, a cohesive understanding of how stressors and resources interact across gender lines, particularly in German higher education, remains underdeveloped. Given the structural transformations and legal reforms affecting mid-level academic staff in Germany, this gap is both timely and pressing.

3. Materials and Methods

This study utilized a standardized online survey to explore the working conditions and job satisfaction of mid-level academic staff across selected disciplines. A total of 10,000 email addresses of mid-level academic staff in economics, law, social sciences, humanities, engineering, mathematics, and natural sciences were randomly collected through the publicly accessible websites of German universities. Potential survey candidates were invited personally vie email to increase the probability of participation (Short et al., 2015). Participants needed to have an employment contract with a university and belong to the professional group of mid-level staff. The absence of an employment contract with a university or belonging to the group of administrative staff or professors led to an exclusion from the survey.

The survey, administered via LimeSurvey between April and May 2023, included the following thematic sections:

• Sociodemographic background including age, gender, marital status, number of children, and responsibilities in caregiving for relatives;
• Employment situation (e.g., contractual working hours, actual weekly workload, and teaching obligations);
• Academic qualifications and career plans (e.g., status of the doctorate and habilitation, participation in structured doctoral programs, long-term academic career aspirations);
• Professional development (e.g., opportunities for further education and conference attendance);
• Workplace culture (e.g., collegial relationships, recognition in academic publishing, gender equity in career advancement);
• Self-efficacy, i.e., the respondent’s belief in their own ability to manage challenges and solve problems;
• Work engagement (e.g., level of inspiration, pride, and perceived meaningfulness derived from the respondent’s academic work);
• Burnout symptoms (e.g., physical, emotional, and cognitive signs of exhaustion and stress);
• Job stressors (e.g., burden experienced from various work-related stressors, including teaching, committee work, grant writing, and financial insecurity);
• Work–life balance (e.g., ability to relax and maintain balance between professional and private life, coping/recreation strategies);
• Family-friendliness and institutional support (e.g., compatibility of family and academic work, support from supervisors).

To ensure the clarity, validity, and reliability of our survey instrument, we conducted a pretest with a sample of 50 academic staff members from our home institution. Based on their feedback, several revisions were made to improve the questionnaire, such as rewording ambiguous or overly complex questions, reducing redundancy among similar items, and refining the sequence of questions to minimize respondent fatigue. Completing the questionnaire took approximately 20 min. Out of the 10,000 people contacted, 1618 individuals started the questionnaire, and 1442 cases were ultimately analyzed. Responses were excluded if less than one-third of the survey was completed (n = 174) or if there was evidence of straight-lining behavior (cf. Kim et al., 2019). This resulted in a response rate of 14.4%.

3.1. Measurement of Variables

In this study, multiple validated instruments were employed to measure the constructs relevant to understanding job satisfaction in academia. The operationalization of each variable is described in detail below, including Cronbach’s alpha of each variable.

Job Satisfaction in Academia was measured using a 10-item scale adapted from Holderberg (2020). The items assess various facets of academic job satisfaction, such as social relations and intrinsic motivation. Example items include “Cooperation with colleagues motivates me” and “If I could choose again, I would choose a job at the university.” Responses were given on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The internal consistency of the scale is acceptable, with a Cronbach’s alpha of 0.735.

Work Dedication was assessed using a five-item scale adapted from the Utrecht Work Engagement Scale by Schaufeli and Bakker (2003). The scale captures the degree of psychological investment and enthusiasm in one’s work. Example items include: “My work is challenging” and “My work inspires me.” Respondents answered using a seven-point frequency scale ranging from 1 (never) to 7 (always). The reliability of the scale is high (Cronbach’s alpha = 0.816).

Self-Efficacy was measured with a ten-item scale adapted from Schwarzer and Jerusalem (1995). The scale assesses individuals’ belief in their ability to cope with various difficult situations. Example items are “I can always manage to solve difficult problems if I try hard enough” and “If someone opposes me, I can find the means and ways to get what I want.” A five-point Likert scale was used ranging from 1 (does not apply at all) to 5 (does fully apply). The scale shows good reliability with a Cronbach’s alpha of 0.847.

Burnout symptoms were measured with a six-item short version adapted from Maslach Burnout Inventory, focusing primarily on emotional exhaustion (Maslach, 1993). An example item is “How often do you feel emotionally exhausted?” Respondents rated each item on a seven-point frequency scale ranging from 1 (never) to 7 (always). The internal consistency of the scale is excellent (Cronbach’s alpha = 0.918).

Perceived Job Insecurity was captured using a three-item scale, also adapted from Holderberg (2020). The items assess different perceived burdens relating to insecure employment conditions, such as “Perceived burden due to unclear continued employment,” “Perceived financial burden due to part-time work,” and “Perceived burden due to fixed-term contract.” A five-point Likert scale from 1 (does not apply at all) to 5 (does fully apply) was used. The reliability of this scale is high (Cronbach’s alpha = 0.829).

Overtime Work was operationalized as the difference between the actual number of hours worked and the number of hours specified in the employment contract. This is a continuous variable derived from self-reported data.

Control Variables: Several dichotomous control variables were included in the analysis:

• Aiming for an Academic Career: Measured by the question “Are you aiming for an academic career?” (1 = yes, 0 = no).
• Children: Assessed with the item “Do you have children?” (1 = yes, 0 = no).
• Postponing the Desire to Have Children: Captured by the question “Have you ever postponed having children for professional reasons?” (1 = yes, 0 = no).
• Structured PhD Program: Measured with the item “Are you doing your PhD as part of a structured doctoral program?” (1 = yes, 0 = no).
• Gender: Assessed with the item “Which gender do you identify with?” (0 = female, 1 = male, 2 = diverse/non-binary)2.
• Age: Included as a continuous variable based on self-reported year of birth.

3.2. Sample Description

The final sample comprised individuals aged 23 to 64 years (M = 34.87, SD = 8.39). Of the respondents, 47.9% self-identified as female, 50.9% as male, and 0.7% as diverse/non-binary. Due to the very small number of participants in the “diverse/non-binary” category, these responses were excluded from subsequent statistical analyses to avoid distortions and ensure adequate statistical power. Regarding relationship status, over three-quarters of respondents were in a partnership (43.7%) or married (31.7%), while 22.7% were single, and a small portion were divorced or widowed. Most respondents were childless (69.4%), while 435 participants reported having children, typically two or fewer (M = 1.8), with a maximum of five. Less than 4% indicated caregiving responsibilities for adult relatives.

The distribution across academic disciplines showed that 28.7% of participants worked in STEM fields, 24.0% in the humanities, 21.2% in law, economics, and social sciences, and 15% in engineering sciences. Disciplinary fields such as medicine, health sciences, sports sciences, agriculture, forestry, nutritional sciences, and arts/cultural sciences were each represented by less than 4% of the sample.

Most respondents (68.3%) worked exclusively as research assistants, while an additional 6.6% combined this role with other job types. About 25% held other roles, including postdocs (8.7%), academic advisors (5.1%), lecturers for special tasks (2.4%), and junior research group leaders (1.2%). Around 1% of respondents were categorized as student research assistants. Overall, 13.3% held more than one type of academic position.

4. Data Analysis

4.1. Research Question 1

To address the first research question—In what ways do work-related variables differ between male and female academics occupying mid-level positions within German academia?—we conducted a series of independent-samples t-tests to examine gender-based differences across key variables, including academic job satisfaction, self-efficacy, work dedication, perceived job insecurity, overtime work, and burnout. The analysis aimed to identify statistically significant disparities in how male and female academics experience and evaluate their working conditions. The following section presents the results organized by variable. Significant gender effects emerged for academic job satisfaction, self-efficacy, overtime work, and burnout (see

Table 1. Descriptive statistics—gender differences in work-related variables (authors’ own work).

Variable	Gender	n		M		SD
Academic Job Satisfaction	f	690		3.03		0.70
	m	734		3.13		0.72
	T = −2.84		p = 0.005 *		Cohen’s d = 0.15
Work Dedication	-
Self-Efficacy	f	690		2.85		0.40
	m	733		2.91		0.42
	T = −2.84		p = 0.005 *		Cohen’s d = 0.14
Burnout	f	690		4.05		1.24
	m	733		3.73		1.29
	T = 4.82		p < 0.001 **		Cohens’s d = 0.26
Perceived Job Insecurity	-
Overtime Work	f	662		7.90		9.22
	m	707		9.64		10.50
	T = −3.27		p = 0.001 **		Cohen’s d = 0.18

* p < 0.05, ** p < 0.01.

).

Men reported significantly higher levels of academic job satisfaction and self-efficacy compared to women. Although the effect sizes were small, these results suggest that male participants tend to evaluate their academic environments more positively and perceive themselves as slightly more competent in managing professional demands.

Regarding overtime work, men reported significantly more overtime hours than women. However, the standard deviations for this variable were notably high in both groups—especially among men—indicating a wide range of individual differences. This suggests that while the mean difference is statistically significant, the variability in overtime work is substantial. In practical terms, this means that although men on average work more overtime, the actual number of overtime hours varies greatly among individuals, and the group means may not fully capture the diversity of working time patterns within each gender group.

In contrast, women reported significantly higher levels of burnout than men, with this variable showing the most pronounced gender difference in the dataset. This points toward a meaningful disparity in the experience of work-related exhaustion, potentially linked to differing work conditions, coping strategies, or role expectations.

No statistically significant gender differences were found for work dedication and perceived job insecurity, indicating comparable levels of engagement and job-related uncertainty across the male and female participants.

4.2. Research Question 2

To examine the second research question—Which organizational and individual factors influence job satisfaction among academic staff?—we conducted multiple linear regression analyses. Independent variables included both organizational factors (e.g., overtime hours, participation in structured PhD programs) and individual characteristics (e.g., gender, age, children, self-efficacy, and burnout symptoms). The dependent variable was job satisfaction in academia. The following section presents the regression results, highlighting the most significant predictors and offering insights into the mechanisms that shape job satisfaction among mid-level academics in German higher education.

All key assumptions for linear regression—linearity, homoscedasticity, independence of errors, and normality of residuals—were thoroughly assessed. Diagnostic plots and statistical tests revealed no violations that would compromise the validity of the models. Multicollinearity was not a concern: all tolerance values exceeded 0.1, and all variance inflation factors (VIFs) were below 5, with most values approaching 1, indicating minimal overlap among independent variables. Accordingly, the models were deemed appropriate for explaining variance in academic job satisfaction.

The analyses were conducted using IBM SPSS Statistics Version 29. Two multiple regression models were estimated to explore factors influencing job satisfaction among mid-level academic staff. Model 1 included a variable measuring perceived job insecurity, while Model 2 excluded this variable due to concerns about scale validation. The comparison between the two models highlights the critical role of job insecurity in shaping job satisfaction. The results of both models are displayed in

Table 2. Results of the regression analysis (authors’ own work).

	Job Satisfaction Among Academic Mid-Level Staff
	Model 1		Model 2
Variable	Std. Error	Estimate	Std. Error	Estimate
(Intercept)	0.139	3.361 ***	0.173	2.705 ***
Gender	0.026	0.019	0.032	0.028
Age	0.002	0.011	0.002	0.082
Aim Academic Career	0.034	0.021	0.043	−0.009
Children	0.032	0.014	0.04	0.006
Postponing the desire to have children	0.028	−0.087 ***	0.034	−0.226 ***
Structured PhD Program	0.031	0.003	0.039	−0.017
Work Dedication	0.013	0.285 ***	0.016	0.243 ***
Self-Efficacy	0.033	0.027	0.041	0.041
Burnout	0.011	−0.175 ***	0.014	−0.274 ***
Perceived Job Insecurity	0.011	−0.559 ***	-	-
Overtime Work	0.001	−0.121 ***	0.002	−0.231 ***
N		1432		1364
R2		0.594		0.348
R2 Adj.		0.590		0.343
F		53.366 ***		72.234 ***
		(df = 11, 1353)		(df = 10, 1354)

*** p < 0.001.

.

Model 1 explains approximately 59.0% of the variance in job satisfaction (adjusted R² = 0.590). The strongest negative predictor was perceived job insecurity (β = −0.559, p < 0.001), underscoring the detrimental impact of precarious employment and funding conditions. Additional significant negative predictors included overtime work (β = −0.121, p < 0.001) and burnout (β = −0.175, p < 0.001). In contrast, work dedication emerged as a robust positive predictor (β = 0.285, p < 0.001).

Interestingly, the variable postponing the desire to have children was negatively associated with job satisfaction (β = −0.087, p < 0.001), whereas having children per se did not exert a significant effect (β = 0.014, p = 0.506). This finding may reflect anticipated conflicts between professional and personal life planning. Other variables—including gender, age, structured PhD program participation, self-efficacy, and academic creer aspirations—did not reach statistical significance.

In Model 2, perceived stress is excluded. The explained variance drops significantly to 34.3% (adjusted R² = 0.343), underscoring the explanatory power of perceived job insecurity. Notably, several variables maintain their significance: postponing the desire to have children (β = −0.226, p < 0.001), overtime work (β = −0.231, p < 0.001), work dedication (β = 0.243, p < 0.001), and burnout (β = −0.274, p < 0.001). Interestingly, the effect of desire to have children becomes even more pronounced in the absence of perceived job insecurity, possibly due to shared variance in concerns over future planning and insecurity. Meanwhile, gender, academic career aspirations, and self-efficacy remain non-significant in both models, suggesting limited direct influence on job satisfaction.

Taken together, the findings suggest that structural stressors—particularly perceived job insecurity and excessive overtime—are key determinants of job satisfaction in academia, while individual-level factors like self-efficacy and gender play a comparatively minor role.

5. Discussion and Conclusions

The present study contributes to the growing body of research on academic job satisfaction by offering a quantitative analysis focused on mid-level academic staff in the German higher education system; an institutional context that remains underrepresented in international research. Our findings both complement and challenge the existing international evidence and highlight important directions for future research. They show that structural stressors, especially working hours and contract insecurity, outweigh individual attributes such as self-efficacy when predicting job satisfaction. This emphasizes the need for policy-level change rather than solely person-centered interventions. To unpack these results in detail, the following section discusses the key predictors of job satisfaction individually, situating each within the context of existing international research and the specific institutional conditions of the German academic system.

Gender (self-reported). Our survey asked, “Which gender do you identify with?”, capturing a social rather than biological construct. Sex assigned at birth was not collected. Because only 0.7% of respondents chose the “diverse/non-binary” option, these cases were excluded from the inferential tests to preserve statistical power—a limitation we revisit below. Across both regression models, the binary gender variable was non-significant, indicating that once key work-related stressors (overtime, burnout, perceived job insecurity) are controlled, male and female mid-level academics do not differ in overall job satisfaction. This result aligns with Mudrák et al. (2018) but diverges from studies showing higher satisfaction for men (Fontinha et al., 2019; Frei & Grund, 2020) or for women (Fetherston et al., 2020). Such inconsistency underscores that gender effects are context-dependent varying with sample composition, contractual status and the set of control variables employed. In our case, once structurally gendered demands (e.g., overtime) are entered, any direct gender effect is reduced in the regression analyses.

Work Dedication emerged as one of the strongest positive predictors of job satisfaction in both models (β = 0.285 *** in Model 1; β = 0.243 *** in Model 2), in line with previous findings that highlight dedication as a core subdimension of work engagement (Demerouti & Bakker, 2006; Schaufeli et al., 2002). Our findings support earlier research across occupational settings, including higher education, which found work engagement and particularly dedication to be robustly linked to satisfaction (Barkhuizen et al., 2014; Mudrák et al., 2018). This reinforces the view that feeling inspired, enthusiastic, and committed to one’s academic work substantially enhances overall well-being. Interestingly, although some studies have pointed to stronger effects for women (Mascarenhas et al., 2022) or higher cynicism and lower dedication among female academics under strain (Bezuidenhout & Cilliers, 2010), our model does not detect any significant gender effect. This may be due to the buffering role of contextual factors such as institutional support, or to the generally high levels of dedication often observed among academics (Mudrák et al., 2018). Despite its empirical relevance, the research specifically focused on dedication, particularly within German higher education, remains sparse, warranting further gender-sensitive and contextualized investigation.

Self-efficacy. Contrary to a large body of research linking self-efficacy to higher satisfaction (Chan et al., 2020; Lesener & Gusy, 2017), the present analysis shows a small, positive but non-significant coefficient. A likely explanation is statistical; self-efficacy shares variance with powerful correlates such as work dedication and burnout (Hall et al., 2019). Once these are partialled out, little unique predictive power remains. In addition, our global self-efficacy measure may be less sensitive to the specialized demands of mid-level academic work; an issue noted in other higher-education studies that favor domain-specific scales. Indirectly, however, self-efficacy may still matter; its theorized buffering role against burnout is compatible with the strong burnout effect observed here.

Burnout emerges as one of the most potent negative predictors of job satisfaction (β = −0.175 in Model 1; −0.274 in Model 2), reinforcing long-standing evidence that emotional exhaustion and cynicism erode well-being in academia (Maslach & Schaufeli, 1993). The coefficient remains robust even after controlling for overtime and job insecurity, underscoring burnout’s status as an independent stress pathway. International studies point to gendered nuances, e.g., higher exhaustion among South-African women (Bezuidenhout & Cilliers, 2010) or elevated feelings of ineffectiveness among Nigerian women (Adekola, 2010) but such patterns were not tested directly here. The German mid-level context, characterized by precarious contracts and high performance expectations, provides fertile ground for the stress–burnout–dissatisfaction chain highlighted in the Job-Demands-Resources framework. While career uncertainty can only be solved structurally, buffers against burnout exist at individual, managerial, and cultural levels. Bogodistov and Moormann (2025) argue that a better person–task fit—achieved through job crafting that lets academics reshape duties around their strengths—adds vital job resources. Such autonomy must be backed by leadership development, as many supervisors enter the role untrained. Supervisor programs that normalize workload and mental-health talks, paired with peer-support or mindfulness initiatives, have lowered faculty burnout (Roeser et al., 2013). At the organizational level, transparent workload and career policies are key. Publishing renewal criteria, setting predictable teaching loads, and coordinating parental-leave cover might reduce the insecurity identified as the main satisfaction drain. Bogodistov and Moormann (2025) further proposed replacing individual publication and funding quotas with department-wide targets, allowing tasks to shift toward each scholar’s talents. These practices will not end fixed-term legislation but can mitigate its impact. Future studies should test whether such fit-focused interventions moderate the stress–burnout–dissatisfaction pathway across genders and disciplines.

Perceived job insecurity shows the largest negative coefficient in Model 1 (β = −0.559 ***), exceeding even burnout. This aligns with the evidence from multiple countries that insecure academic employment undermines well-being, particularly for women (Corvino et al., 2022; Mudrák et al., 2018). In Germany, fixed-term hiring governed by the Act on Fixed-Term Employment in Academia (Wissenschaftszeitvertragsgesetz) leaves up to 90% of mid-level staff on time-limited contracts with no automatic path to permanence (Ambrasat, 2019; Briedis et al., 2020). Beyond contract length, opaque career trajectories (unclear promotion milestones, limited tenure quotas, and scarce permanent alternatives outside professorships) fuel chronic uncertainty. Such blurred career pathways amplify subjective insecurity beyond the legal term limit itself, fostering a climate of continual job-searching and delayed life planning. Accordingly, any policy aimed at boosting job satisfaction must tackle both contract duration (e.g., longer initial terms, transparent tenure criteria) and the wider architecture of academic career trajectories: clarifying promotion milestones, tenure quotas, and alternative permanent roles other than professorship within the university system.

Overtime work. Consistent with the findings that excessive hours and intrusive work thoughts harm well-being (Fetherston et al., 2020), overtime is a strong negative predictor (β = −0.121 *** in Model 1; −0.231 *** in Model 2) for job satisfaction. This suggests that overtime exerts strain over and above burnout or low resources. Prior German evidence indicates that men accumulate more overtime, whereas women, especially those with caregiving duties, experience greater cognitive stress at comparable or lower hour totals (Geier et al., 2025). Although gender interactions were not modelled here, the general negative association indicates that long hours remain a universal detractor of satisfaction. Germany’s current debate on time-tracking for academic mid-level staff could help visualize the full workload, but technical monitoring must be paired with clear task definitions and a workplace culture that discourages chronic overtime (Frei & Grund, 2020; Geier et al., 2025). Policies that redistribute administrative duties and recognize reasonable working hours as a sign of professional effectiveness would reinforce such a culture.

In sum, German mid-level academics occupy a structurally vulnerable group: contracts are overwhelmingly temporary, teaching and administrative loads are rising, and clear tenure pathways are scarce. Recent legislative changes (e.g., the WissZeitVG) and equality initiatives (e.g., “Professorinnenprogramm”, a program to promote female professorships) have sought to improve conditions, yet the present results indicate that systemic stressors, foremost job insecurity, overtime, and burnout, continue to overshadow personal resources such as self-efficacy.

The strong and consistent negative association between postponed parenthood and job satisfaction warrants further investigation, as it may reflect broader systemic incompatibilities between academic career trajectories and family planning. The regression analyses highlight three principal levers of job satisfaction among German mid-level academics: perceived job insecurity, burnout, and overtime work. Each represents a structural rather than individual deficit, suggesting that organizational and policy reforms, not merely individual coping strategies, are requisite for sustainable well-being. Self-efficacy, while theoretically salient, exerts little direct influence once these structural forces are accounted for. Finally, in this controlled model, gender does not predict satisfaction, implying that observed gender gaps in other contexts may largely reflect unequal exposure to the very demands and insecurities identified here. In addition to contract and workload reforms, universities should adopt family-inclusive and gender-diverse support policies that recognize stressors such as postponed parenthood across all identities. Future studies ought to disaggregate sex—the biological capacity for pregnancy, childbirth, and related leave—from gender, the social roles, and expectations that shape service load or caregiving norms, so that biological and sociocultural pathways to satisfaction can be distinguished. The adequate sampling of non-binary academics and intersectional variables (e.g., caregiving status) will further clarify which groups are most affected and guide targeted interventions. Addressing contract insecurity, regulating workload, and instituting burnout prevention programs thus appear crucial for fostering a more satisfied and resilient mid-level academic workforce in Germany and, by extension, comparable higher-education systems worldwide.

6. Limitations and Future Research

While this study provides valuable insights into job satisfaction among mid-level academic staff in Germany, several limitations must be acknowledged.

First, the national focus limits generalizability to other higher education systems with different legal and career structures (e.g., tenure-track models). Comparative international studies are needed to assess whether patterns like the impact of job insecurity or postponed parenthood are generalized across academic cultures. Second, although diverse disciplines are included, the sample does not fully represent the German academic landscape. Fields such as law, medicine, or the arts may be underrepresented. Given the disciplinary variation in academic cultures, future studies should model disciplinary affiliation more explicitly or conduct field-specific analyses.

The absence of a formalized theoretical framework constitutes another limitation of this study. Instead, we based our empirical approach on an extensive review of the relevant literature that guided our regression analysis. The chosen analytical strategy allowed us to explore the research questions in a robust and transparent way, resulting in a model that explains a substantial proportion of the variance in the outcome. Future research could use more theory-based approaches and model-based testing in order to develop more generalizable conclusions and strengthen the theoretical grounding of findings. Despite an exploratory approach, the findings align with the J-D-R model and burnout theory. Thus, our results can be used to support the aforementioned practical implications and serve as a foundation for future research.

As in many other studies, the cross-sectional design precludes causal inference. Longitudinal studies are necessary to track how stressors and protective factors interact over time, and how career transitions (e.g., parenthood, contract changes) influence satisfaction. Also, self-selection bias remains possible, as participation was voluntary and online. Finally, gender was included but restricted to binary categories due to the low response rates from non-binary individuals. This reflects broader challenges in inclusion and visibility in academic research. Future work should adopt intersectional and inclusive designs to explore how gender interacts with other identity dimensions (e.g., caregiving, disability, migration status).